With the exponential enhancement of manufacturing effectivity and the structural transformation of the software of manufacturing elements induced by using the new spherical of scientific and technological revolution, the sample of comparative gain of manufacturing elements in the worldwide manufacturing enterprise has passed through profound changes, progressively forming a transformation from the industrial chain to the price chain, therefore promoting the transformation of manufacturing mode and aid utilization mode. Based on the applicable theories of wise manufacturing, this paper proposes an clever manufacturing functionality assessment and prediction mannequin primarily based on utilized mathematical modelling, which integrates the contrast facts of agencies into a database, establishes the corresponding time sequence relationship of the comparison data, and establishes a relationship with the corresponding organization attributes, fashions the assessment data, and combines them into a total prediction mannequin through built-in learning. By functionality of migration learning, the prediction model applicable to large statistics is migrated to small information to recognize the distinction of the equal type alternatively special entities. Input the comparison facts of the organization to be evaluated into the prediction mannequin to reap the shrewd manufacturing functionality maturity fee of the enterprise. It realizes the automated comparison of shrewd manufacturing capability, saves human cost, and improves the accuracy of evaluation. It solves the troubles that manufacturing companies have no longer deep hold close and understanding of smart manufacturing, inaccurate identification of their very personal smart manufacturing enhancement stage, and unscientific self-evaluation and diagnosis.